# [PATCH] add -p: avoid use of undefined $key when ReadKey -> EOF

4 messages from 2021-11-28 to 2021-11-28. Participants: Carlo Marcelo Arenas Belón, Junio C Hamano, Carlo Arenas.
Thread: https://gitlist.dev/t/56992

## Carlo Marcelo Arenas Belón, 2021-11-28 17:49

Subject: [PATCH] add -p: avoid use of undefined $key when ReadKey -> EOF
Message-ID: <20211128174903.5623-1-carenas@gmail.com>
URL: https://gitlist.dev/e/20211128174903.5623-1-carenas%40gmail.com

```
b5cc003253 (add -i: ignore terminal escape sequences, 2011-05-17)
add an additional check to the original code to better handle keys
for escape sequences, but failed to account for the possibility
the first ReadKey call returned undef (ex: stdin closes) and that
was being handled fine by the original code in ca6ac7f135 (add -p:
prompt for single characters, 2009-02-05)

Add a test for undefined and encapsulate the loop and the original
print that relied on it within it.

After this, the following command (in a suitable repository state)
wouldn't print any error:

  $ git -c interactive.singleKey add -p </dev/null

Signed-off-by: Carlo Marcelo Arenas Belón <carenas@gmail.com>
---
 git-add--interactive.perl | 16 +++++++++-------
 1 file changed, 9 insertions(+), 7 deletions(-)

diff --git a/git-add--interactive.perl b/git-add--interactive.perl
index bc3a1e8eff..95887fd8e5 100755
--- a/git-add--interactive.perl
+++ b/git-add--interactive.perl
@@ -1175,15 +1175,17 @@ sub prompt_single_character {
 		ReadMode 'cbreak';
 		my $key = ReadKey 0;
 		ReadMode 'restore';
-		if ($use_termcap and $key eq "\e") {
-			while (!defined $term_escapes{$key}) {
-				my $next = ReadKey 0.5;
-				last if (!defined $next);
-				$key .= $next;
+		if (defined $key) {
+			if ($use_termcap and $key eq "\e") {
+				while (!defined $term_escapes{$key}) {
+					my $next = ReadKey 0.5;
+					last if (!defined $next);
+					$key .= $next;
+				}
+				$key =~ s/\e/^[/;
 			}
-			$key =~ s/\e/^[/;
+			print "$key";
 		}
-		print "$key" if defined $key;
 		print "\n";
 		return $key;
 	} else {
-- 
2.34.0.352.g07dee3c5e1


```

## Junio C Hamano, 2021-11-28 19:44

Subject: Re: [PATCH] add -p: avoid use of undefined $key when ReadKey -> EOF
Message-ID: <xmqq8rx85ala.fsf@gitster.g>
URL: https://gitlist.dev/e/xmqq8rx85ala.fsf%40gitster.g
In-Reply-To: <20211128174903.5623-1-carenas@gmail.com>

```
Carlo Marcelo Arenas Belón  <carenas@gmail.com> writes:

> b5cc003253 (add -i: ignore terminal escape sequences, 2011-05-17)
> add an additional check to the original code to better handle keys
> for escape sequences, but failed to account for the possibility
> the first ReadKey call returned undef (ex: stdin closes) and that
> was being handled fine by the original code in ca6ac7f135 (add -p:
> prompt for single characters, 2009-02-05)
>
> Add a test for undefined and encapsulate the loop and the original
> print that relied on it within it.

I wondered where a change to t/ directory was, but we are not
talking about that kind of test ;-)  OK.

> After this, the following command (in a suitable repository state)
> wouldn't print any error:
>
>   $ git -c interactive.singleKey add -p </dev/null
>
> Signed-off-by: Carlo Marcelo Arenas Belón <carenas@gmail.com>
> ---
>  git-add--interactive.perl | 16 +++++++++-------
>  1 file changed, 9 insertions(+), 7 deletions(-)
>
> diff --git a/git-add--interactive.perl b/git-add--interactive.perl
> index bc3a1e8eff..95887fd8e5 100755
> --- a/git-add--interactive.perl
> +++ b/git-add--interactive.perl
> @@ -1175,15 +1175,17 @@ sub prompt_single_character {
>  		ReadMode 'cbreak';
>  		my $key = ReadKey 0;
>  		ReadMode 'restore';
> -		if ($use_termcap and $key eq "\e") {
> -			while (!defined $term_escapes{$key}) {
> -				my $next = ReadKey 0.5;
> -				last if (!defined $next);
> -				$key .= $next;
> +		if (defined $key) {
> +			if ($use_termcap and $key eq "\e") {
> +				while (!defined $term_escapes{$key}) {
> +					my $next = ReadKey 0.5;
> +					last if (!defined $next);
> +					$key .= $next;
> +				}
> +				$key =~ s/\e/^[/;
>  			}
> -			$key =~ s/\e/^[/;
> +			print "$key";
>  		}
> -		print "$key" if defined $key;

This was from the very original and was a strong clue that $key
could be undef; well spotted.

>  		print "\n";
>  		return $key;
>  	} else {

Essentially, the code added by b5cc0032 (add -i: ignore terminal
escape sequences, 2011-05-17) wanted to say

    The original code

    - called ReadKey and took its output in $key, 
    - echoed it if it was defined (otherwise $key wasn't echoed),
    - emitted a line feed,
    - and returned $key.

    We deal with the case where a single keystroke of some keys is
    delivered to us as an escape sequence of multiple "key"s from
    ReadKey, and the way we do so is to replace that single ReadKey
    with the new code that accumulates these multiple "key"s in
    $key.

I am undecided, but the minimum patch below seems to makes the above
intention in the updated code clearer.

diff --git i/git-add--interactive.perl w/git-add--interactive.perl
index bc3a1e8eff..c459c675e7 100755
--- i/git-add--interactive.perl
+++ w/git-add--interactive.perl
@@ -1175,7 +1175,7 @@ sub prompt_single_character {
 		ReadMode 'cbreak';
 		my $key = ReadKey 0;
 		ReadMode 'restore';
-		if ($use_termcap and $key eq "\e") {
+		if (defined $key && $use_termcap and $key eq "\e") {
 			while (!defined $term_escapes{$key}) {
 				my $next = ReadKey 0.5;
 				last if (!defined $next);

```

## Carlo Arenas, 2021-11-28 20:52

Subject: Re: [PATCH] add -p: avoid use of undefined $key when ReadKey -> EOF
Message-ID: <CAPUEspjo+cKLnE+MoABp_NnGDCCz8fL=pJj3ZNsKmCVetUn1jA@mail.gmail.com>
URL: https://gitlist.dev/e/CAPUEspjo%2BcKLnE%2BMoABp_NnGDCCz8fL%3DpJj3ZNsKmCVetUn1jA%40mail.gmail.com
In-Reply-To: <xmqq8rx85ala.fsf@gitster.g>

```
On Sun, Nov 28, 2021 at 11:44 AM Junio C Hamano <gitster@pobox.com> wrote:
> Carlo Marcelo Arenas Belón  <carenas@gmail.com> writes:
> > Add a test for undefined and encapsulate the loop and the original
> > print that relied on it within it.
>
> I wondered where a change to t/ directory was, but we are not
> talking about that kind of test ;-)  OK.

Will change the wording in v2 with something less ambiguous like
"Check for undefined.."

> I am undecided, but the minimum patch below seems to makes the above
> intention in the updated code clearer.
>
> diff --git i/git-add--interactive.perl w/git-add--interactive.perl
> index bc3a1e8eff..c459c675e7 100755
> --- i/git-add--interactive.perl
> +++ w/git-add--interactive.perl
> @@ -1175,7 +1175,7 @@ sub prompt_single_character {
>                 ReadMode 'cbreak';
>                 my $key = ReadKey 0;
>                 ReadMode 'restore';
> -               if ($use_termcap and $key eq "\e") {
> +               if (defined $key && $use_termcap and $key eq "\e") {
>                         while (!defined $term_escapes{$key}) {
>                                 my $next = ReadKey 0.5;
>                                 last if (!defined $next);

The issue I had with that was that the order of the checks is slightly
awkward and inefficient (ex: $use_termcap = false probably should
short circuit, and both checks for $key make more sense next to each
other) and doing it outside avoids the extra !defined check for print.

It would also look cleaner IMHO if the expression would be rewritten
to use '&&' instead of 'and', which smells like unnecessary
refactoring that I intentionally tried to avoid by creating a
precondition instead so `diff -w` is really small and straight
forward.

I have no strong opinions eitherway, so let me know what you would
prefer and will get a v2 with it; thanks for reviewing.

Carlo

PS. list email is for whatever reason taking a long time to arrive
from vger, hopefully not many are affected, but I might be slow to
respond

```

## Junio C Hamano, 2021-11-28 23:15

Subject: Re: [PATCH] add -p: avoid use of undefined $key when ReadKey -> EOF
Message-ID: <xmqq4k7v6fdn.fsf@gitster.g>
URL: https://gitlist.dev/e/xmqq4k7v6fdn.fsf%40gitster.g
In-Reply-To: <CAPUEspjo+cKLnE+MoABp_NnGDCCz8fL=pJj3ZNsKmCVetUn1jA@mail.gmail.com>

```
Carlo Arenas <carenas@gmail.com> writes:

> I have no strong opinions eitherway, so let me know what you would
> prefer and will get a v2 with it; thanks for reviewing.

I have no strong preference either way, so I just queued what we
reviewed as-is.

Thanks for the fix.

```
